PFAS
PFAS is modelled with the same build-up and wash-off machinery as any other pollutant, but almost every default that suits TSS is wrong for it. The differences are not cosmetic: get them wrong and the model will tell you a treatment train works when it does nothing at all.
What makes it different
Section titled “What makes it different”It is dissolved and mobile. PFAS stays in the water column. It does not bind to sediment, so it does not travel with TSS and is not removed with TSS.
It does not decay. The “forever” in forever chemicals. Decay is set to zero rather than to a long half-life.
It is measured at trace concentrations. Reported in ng/L, entered in the model as µg/L.
8.5 means 8,500 ng/L. This is the single easiest place to be out by a factor of a thousand.
Because dry-weather PFAS build-up is still poorly characterised, it is set up as an EMC wash-off - a fixed event-mean concentration per land use - rather than a build-up curve that would imply knowledge nobody has.
Setting it up
Section titled “Setting it up”-
On the Setup tab, choose a PFAS constituent from the violet PFAS row: Total PFAS, PFOA or PFOS. It is seeded as a standalone pollutant, never as a TSS co-pollutant.
-
Check the event-mean concentrations in the PFAS table. The defaults are a starting point, not a recommendation:
Land use Default EMC (µg/L) Residential 0.05 Commercial 0.25 Industrial 8.5 A trace atmospheric background is also applied through rainfall.
-
Mark any hotspots. The Industrial default represents a mixed industrial zone. Airports, fire-training grounds and other AFFF sites run far higher - 5 to 50+ µg/L. Use Mark as AFFF hotspot to assign those subcatchments to a high-PFAS land use.
-
Run, then read the outfall pollutograph. The peak is shown against an EPA drinking-water reference line (the 4 ng/L MCL for PFOA and PFOS, the legacy 70 ng/L health advisory for Total PFAS) - a screening comparison, not a stormwater limit.
The part that catches people out
Section titled “The part that catches people out”Real capture needs adsorptive media - colloidal activated carbon, granular activated carbon, or ion exchange. Model it as a high fixed percentage removal for the PFAS pollutant at the treatment node, on the Controls tab. See biofiltration BMPs for the one-step version.
Verified on the Watertown model: a 90 percent node removal cuts the outfall PFAS peak from 4.4 to 0.44 µg/L, while a TSS-only settling treatment leaves it unchanged.
Reporting it honestly
Section titled “Reporting it honestly”The uncertainty warning that applies to all quality modelling applies doubly here. EMCs vary by orders of magnitude between sites, and a model seeded from defaults is a screening tool for comparing options, not a prediction of concentration. Say which EMCs you used and where they came from.